Impact of GenAI on ideation and innovation team performance

This dataset accompanies the study titled "The Impact of Generative Artificial Intelligence on Ideation and the Performance of Innovation Teams" by Michael Gindert and Marvin Lutz Müller. The research investigates how a Generative AI (GenAI)-augmented ideation tool affects ideation quality, efficiency, and team dynamics within a structured innovation process. Utilizing the Knowledge Spillover Theory of Entrepreneurship (KSTE) as a framework, the data highlights the influence of AI on knowledge generation, transfer, and application during the ideation phase. A framed field experiment has been conducted. Two groups (experiment and control) had to ideate on two destinct innovation tasks. The experiment group used an AI augmented ideation tool, that has been developed for the study.

Hypotheses: The dataset supports the testing of six main hypotheses concerning the application of large language models (LLMs) during ideation in team settings:

H1: AI-augmented teams will produce higher-quality ideas due to enhanced knowledge spillover.
H2: AI support will accelerate the ideation process.
H3: Teams using AI will exhibit greater efficiency in idea generation.
H4: AI-assisted teams will produce a more diverse set of ideas, in terms of novelty and feasibility.
H5: AI-enhanced ideation increases the likelihood of generating revolutionary ideas.
H6: AI-supported ideation will positively impact team satisfaction and engagement.
Data Highlights and Findings:

Idea Quality: The dataset includes quality assessments of ideas based on originality, feasibility, and clarity, revealing that AI-augmented teams consistently generated higher-quality ideas across various dimensions.
Process Speed and Efficiency: Time measurements show that teams with AI support completed ideation tasks significantly faster than the control group, reducing ideation time by up to 30%.
Diversity and Novelty: Evaluations indicate that ideas produced with AI support exhibited greater diversity in both approach and innovation potential, contributing to more revolutionary solutions, especially for complex problems in healthcare and automotive sectors.
Team Satisfaction: Survey data suggest that team members in the AI-augmented condition reported higher levels of engagement and satisfaction, with a 23.3% increase in positive feedback compared to the control group.

本数据集伴随由Michael Gindert与Marvin Lutz Müller所著之研究论文《生成式人工智能对创新团队理念生成及绩效的影响》，旨在探讨生成式人工智能（GenAI）增强的理念生成工具如何影响理念生成的质量、效率以及团队在结构化创新过程中的动态。研究以创业知识溢出理论（KSTE）为框架，突显了人工智能在理念生成阶段对知识生成、转移和应用的影响。通过一项框定现场实验，分为实验组和对照组的两个团队分别对两个不同的创新任务进行理念生成。实验组使用了一款专为研究开发的AI增强理念生成工具。

假设：本数据集支持对六个主要假设的测试，这些假设涉及在大语言模型（LLM）支持下团队理念生成的应用：

H1：由于知识溢出的增强，AI增强的团队将产生更高质量的想法。
H2：AI支持将加速理念生成过程。
H3：使用AI的团队在理念生成中将展现更高的效率。
H4：AI辅助的团队将产生更多样化的想法，在新颖性和可行性方面有所体现。
H5：AI增强的理念生成将提高产生革命性想法的可能性。
H6：AI支持的理念生成将对团队满意度和参与度产生积极影响。

数据亮点与发现：
理念质量：数据集包括基于原创性、可行性和清晰度的理念质量评估，显示AI增强的团队在多个维度上持续产生高质量的理念。
过程速度与效率：时间测量表明，获得AI支持的团队完成理念生成任务的速度比对照组显著更快，理念生成时间减少了高达30%。
多样性与新颖性：评估表明，获得AI支持的生成的想法在方法和创新潜力方面表现出更大的多样性，有助于产生更多革命性解决方案，特别是在医疗和汽车行业的复杂问题上。
团队满意度：调查数据显示，AI增强条件下的团队成员报告了更高的参与度和满意度，与对照组相比，正面反馈增加了23.3%。